wigoloMCP Server

Released Jul 19, 2026

v0.2.1

AGPL-3.0-only

Summary

Local-first web MCP server for AI agents — search, fetch, crawl, extract and deep research with no API keys, no cloud round-trip and no per-query bill.

Features

Multi-engine search across 18 adapters with rank fusion and ML reranking
Tiered fetch that escalates to a headless browser only when a page needs it
Multi-page crawl by BFS, DFS or sitemap with rate limits and robots.txt respect
Structured extraction: tables, JSON-LD, named schemas or a custom JSON Schema
Deep research tool that decomposes a question and synthesizes a cited report
diff and watch report page changes since last visit, with webhook delivery
Core tools need no API keys; cached data stays in ~/.wigolo
One-command setup that also wires Claude Code, Cursor, Codex and six other agents

Installation

Set up this MCP server in your favorite AI agent — copy a ready-made configuration below.

Any MCP-compatible agent

Most agents (Claude, Cursor, Windsurf, VS Code, and more) read a standard mcpServers configuration.

Open your agent's MCP configuration file.
Merge the snippet below into it.
Restart the agent — the "wigolo" tools will be available.

{
  "mcpServers": {
    "wigolo": {
      "command": "npx",
      "args": [
        "-y",
        "wigolo"
      ]
    }
  }
}

Claude Code

Add the server with a single CLI command.

claude mcp add wigolo -- npx -y wigolo

Claude Desktop

Add the server to claude_desktop_config.json (Settings → Developer → Edit Config).

{
  "mcpServers": {
    "wigolo": {
      "command": "npx",
      "args": [
        "-y",
        "wigolo"
      ]
    }
  }
}

Cursor

{
  "command": "npx",
  "args": [
    "-y",
    "wigolo"
  ]
}

Windsurf

{
  "command": "npx",
  "args": [
    "-y",
    "wigolo"
  ]
}

VS Code

{
  "command": "npx",
  "args": [
    "-y",
    "wigolo"
  ]
}

Set up the local engine and wire your agents in one command

{
  "command": "npx",
  "args": [
    "wigolo",
    "init",
    "--agents=claude-code,cursor"
  ]
}

Health-check the local engine and downloaded components

{
  "command": "npx",
  "args": [
    "wigolo",
    "doctor"
  ]
}

Description

wigolo gives a coding agent a single surface for everything web-related: search, fetch, crawl, extract, cache, find-similar, research and autonomous gather loops. The core tools need no API keys and nothing they touch leaves ~/.wigolo/, which makes it a practical alternative to metered search-and-scrape APIs for agents that browse heavily.

The toolset is unusually broad for a single server. search runs multi-engine web search across 18 direct adapters with rank fusion, ML reranking and an explainable per-result score, and accepts an array of queries for parallel breadth. fetch loads a URL through a tiered router that escalates from plain HTTP to a headless browser only when it meets an anti-bot challenge or an SPA shell, returning clean markdown with metadata and links; it also handles PDFs, single-heading sections, authenticated sessions and page actions such as click, type, scroll and screenshot. crawl walks multiple pages by BFS, DFS or sitemap with per-domain rate limits, robots.txt compliance and boilerplate dedup. extract pulls tables, metadata, JSON-LD, brand identity, named schemas or any custom JSON Schema, while diff and watch report what changed on a page since the last visit and can deliver changes to a webhook.

Setup is one command: npx wigolo init provisions the local engine, and --agents=claude-code,cursor also writes the MCP config for the agents you name. Supported targets include Claude Code, Cursor, Codex, Gemini CLI, OpenCode, VS Code, Windsurf, Zed and Antigravity, plus any MCP client, LangChain, CrewAI, LlamaIndex, the Vercel AI SDK, n8n or plain REST. Every tool also runs from the terminal, an interactive NDJSON shell and the SDKs.

Requires Node 20 or newer and roughly 1.5 GB of free disk on macOS, Linux or Windows. AGPL-3.0-only and currently in public beta.

Related MCP Servers

MCP: jevcrawl

by Rayen Mlayeh

Goal-directed web crawler MCP server that returns only the passages answering your agent's question, verbatim and with sources.

Web & Search

MCP: SerpApi MCP Server

by SerpApi

Official SerpApi MCP server: real-time Google, Bing, YouTube and other search results as structured JSON or token-saving markdown.

Web & Search

MCP: Crawlbase MCP

by Crawlbase

Official Crawlbase MCP server giving agents live web access — fetch any URL as raw HTML, clean Markdown or a screenshot, with JS rendering and proxy rotation handled for you.

Web & Search

MCP: Exa Search

by Exa

Real-time web search and crawling built for AI — neural search with configurable tools and live content retrieval.

Web & Search

Browse all MCP servers →